garycase Posted November 7, 2016 Share Posted November 7, 2016 There have been various improvements to the shutdown processes, so it would be useful to hear if there are still shutdown issues, using this latest version, especially from those who have had issues with shutting down in the past. What information would be useful? Mine seems to hang since I no longer have Powerdown. Still on 6.2.2 while I wait for a parity rebuild and check to finish. Are any logs/diagnostics preserved now on reboot? First I'd upgrade to 6.2.4. Then try a shutdown and see if it works okay => if it hangs, it's now supposed to generate a log file that will provide some feedback for LimeTech to review so Tom can continue to improve the shutdown process. LimeTech's goal has been to eliminate the need for the PowerDown plugin ... and dlandon is no longer supporting versions beyond 6.2 You may want to read this thread: https://lime-technology.com/forum/index.php?topic=51983.90 Quote Link to comment
wgstarks Posted November 8, 2016 Share Posted November 8, 2016 There have been various improvements to the shutdown processes, so it would be useful to hear if there are still shutdown issues, using this latest version, especially from those who have had issues with shutting down in the past. What information would be useful? Mine seems to hang since I no longer have Powerdown. Still on 6.2.2 while I wait for a parity rebuild and check to finish. Are any logs/diagnostics preserved now on reboot? First I'd upgrade to 6.2.4. Then try a shutdown and see if it works okay => if it hangs, it's now supposed to generate a log file that will provide some feedback for LimeTech to review so Tom can continue to improve the shutdown process. LimeTech's goal has been to eliminate the need for the PowerDown plugin ... and dlandon is no longer supporting versions beyond 6.2 You may want to read this thread: https://lime-technology.com/forum/index.php?topic=51983.90 Ok. Updated to 6.2.4 and rebooted with dockers stopped and array offline. No problems. Rebooted again using the GUI Power Button. Dockers running. Array started. Failed to shutdown properly (hung). Had to run reboot from telnet. The array was stopped after reboot completed even though it was running when I first initiated the reboot. I'm guessing it had stopped by the time I telneted in??? Diagnostics attached but I don't see anything in there from before the reboot. If there's some other file just let me know. brunnhilde-diagnostics-20161107-2138.zip Quote Link to comment
JonathanM Posted November 8, 2016 Share Posted November 8, 2016 Ok. Updated to 6.2.4 and rebooted with dockers stopped and array offline. No problems. Rebooted again using the GUI Power Button. Dockers running. Array started. Failed to shutdown properly (hung). How long did you wait? IIRC, the timeout process is fairly long, and the GUI won't respond until it's done. Quote Link to comment
archedraft Posted November 8, 2016 Share Posted November 8, 2016 Ok. Updated to 6.2.4 and rebooted with dockers stopped and array offline. No problems. Rebooted again using the GUI Power Button. Dockers running. Array started. Failed to shutdown properly (hung). How long did you wait? IIRC, the timeout process is fairly long, and the GUI won't respond until it's done. FWIW, the Power Down Button plugin is no longer working for me since any of the 6.2+ updates. I think it's because I no longer have powerdown plugin. I was wondering if that power button plugin is sending the powerdown -r command which doesn't exist anymore? I have had very little time to do much of anything with unRAID other than try and keep up with this crazy amount of updates (kudos!). Quote Link to comment
wgstarks Posted November 8, 2016 Share Posted November 8, 2016 Ok. Updated to 6.2.4 and rebooted with dockers stopped and array offline. No problems. Rebooted again using the GUI Power Button. Dockers running. Array started. Failed to shutdown properly (hung). How long did you wait? IIRC, the timeout process is fairly long, and the GUI won't respond until it's done. It was 5 minutes when I decided to telnet in. Probably another minute or so to log in and run the reboot command. Quote Link to comment
jtech007 Posted November 8, 2016 Share Posted November 8, 2016 Upgraded from 6.2.3 with no issues. Stopped docker and Win10 VM before doing so, seemed to be faster/better than prior updates with those services running. Quote Link to comment
MTA99 Posted November 8, 2016 Share Posted November 8, 2016 Updated 6.2.0 >> 6.2.4 No issues Quote Link to comment
danioj Posted November 8, 2016 Share Posted November 8, 2016 Upgraded again. No issues. This new focus on Security Vulnerabilities is EXCELLENT LT! Awesome! Quote Link to comment
betaman Posted November 8, 2016 Share Posted November 8, 2016 There have been various improvements to the shutdown processes, so it would be useful to hear if there are still shutdown issues, using this latest version, especially from those who have had issues with shutting down in the past. What information would be useful? Mine seems to hang since I no longer have Powerdown. Still on 6.2.2 while I wait for a parity rebuild and check to finish. Are any logs/diagnostics preserved now on reboot? First I'd upgrade to 6.2.4. Then try a shutdown and see if it works okay => if it hangs, it's now supposed to generate a log file that will provide some feedback for LimeTech to review so Tom can continue to improve the shutdown process. LimeTech's goal has been to eliminate the need for the PowerDown plugin ... and dlandon is no longer supporting versions beyond 6.2 You may want to read this thread: https://lime-technology.com/forum/index.php?topic=51983.90 Thanks for the link. I should qualify my original post by saying the Powerdown plugin was active on my server during these delayed shutdowns. I was aware about some issues with this plugin and 6.2 but I wasn't entirely sure if we should keep using it or not? Based on what I read in the link you provided, should I remove it now in 6.2.4 or should I wait for 6.3 final (no plan to use an rc or beta at this point)? Quote Link to comment
garycase Posted November 8, 2016 Share Posted November 8, 2016 I'd leave Powerdown installed until 6.3, as there are still some "tweaks" Tom is adding to the process. Quote Link to comment
TUMS Posted November 8, 2016 Share Posted November 8, 2016 Updated fine 6.2.1-->6.2.4. Doing a parity check now. Quote Link to comment
kizer Posted November 8, 2016 Share Posted November 8, 2016 From 6.1.9 -> 6.2.4 with zero noticeable issues Quote Link to comment
ryoko227 Posted November 9, 2016 Share Posted November 9, 2016 6.2.3 to 6.2.4 on two machines, no issues so far. Quote Link to comment
hhs99 Posted November 9, 2016 Share Posted November 9, 2016 did this fix where some VM's will hang on shutdown with Nvidia Cards passed through? Quote Link to comment
chris1259 Posted November 10, 2016 Share Posted November 10, 2016 I had an issue today. About 10hrs after i upgraded to v6.2.4 from v6.2.3. First the Windows 10 VM with Nvidia passthrough fell off the network. Then the mouse started to act erratically. I rebooted the VM from within Windows but it hung on Restarting for over 30mins. I tried to connect to the server via web browser but i could not proceed past the login. Then I SSL into the server and tried to reboot it from the command line. It said it worked but it didn't. The last step was pulling the power chord. The server and Windows 10 VM are now working again. I notice every few upgrades, the VM fails to boot properly on the first boot after an upgrade. Then the issue resolves itself. Next time it happens i will post a log file. Quote Link to comment
spencers Posted November 10, 2016 Share Posted November 10, 2016 I just experienced my first hiccup with my Win10 VM (nvidia 1070 passed through) in a long long time. Only thing I did recently was update to 6.2.4. I rebooted the VM from the Windows GUI and after logging in, the screen just goes black. So then I tried a shut down of the VM from the unRAID GUI and it wouldn't shut down. I had to force stop the VM. Upon reboot of the VM, it goes to the same black screen after logging in. I'm rebooting the unRAID server to see if that does anything. edit: Now my VM is just hanging at logging on. This was after seeing two blue screens: one about trying to write memory to read only followed by a system exception not handled. I'm stuck with a non-working VM now! edit2: Luckily I had a backup of the VM on the array from two weeks ago. I pointed the VM to use that disk and it boot fine. Do qcow2 files go bad? It's a 300GB whopper! Quote Link to comment
SCSI Posted November 10, 2016 Share Posted November 10, 2016 Upgraded with no issues. Thanks! I don't like this rapid release cycle since I can't get the uptime on my server as long as it used to before. /s Quote Link to comment
interwebtech Posted November 10, 2016 Share Posted November 10, 2016 the issue with Docker restart causing the array to wake up (and delay waiting for disks to spin up causes docker restart to throw an error, timeout likely) is still alive and in this new version too. http://lime-technology.com/forum/index.php?topic=53340.msg514465#msg514465 Quote Link to comment
spencers Posted November 10, 2016 Share Posted November 10, 2016 After copying the VM disk from my array back to my SSD, the VM boots but is stuck at the windows 10 logo and spinning circle. I've turned off the GTX 1070 passthrough and just using VNC for simplicity's sake. I'm baffled as to what is wrong! The VM disk on the array boot just fine last night as a test. Why on earth won't it boot now after I copied it with midnight commander to my SSD? Quote Link to comment
Frank1940 Posted November 10, 2016 Share Posted November 10, 2016 After copying the VM disk from my array back to my SSD, the VM boots but is stuck at the windows 10 logo and spinning circle. I've turned off the GTX 1070 passthrough and just using VNC for simplicity's sake. I'm baffled as to what is wrong! The VM disk on the array boot just fine last night as a test. Why on earth won't it boot now after I copied it with midnight commander to my SSD? Make sure that you didn't miss any system or hidden files when you did the copy. (Some file managers don't automatically display/transfer those files.) Quote Link to comment
jonp Posted November 10, 2016 Share Posted November 10, 2016 I had an issue today. About 10hrs after i upgraded to v6.2.4 from v6.2.3. First the Windows 10 VM with Nvidia passthrough fell off the network. Then the mouse started to act erratically. I rebooted the VM from within Windows but it hung on Restarting for over 30mins. I tried to connect to the server via web browser but i could not proceed past the login. Then I SSL into the server and tried to reboot it from the command line. It said it worked but it didn't. The last step was pulling the power chord. The server and Windows 10 VM are now working again. I notice every few upgrades, the VM fails to boot properly on the first boot after an upgrade. Then the issue resolves itself. Next time it happens i will post a log file. Hi Chris, There isn't really enough information in your post to diagnose what's going on. First and foremost, how are you assigning a USB mouse to the VM? Are you passing through an entire USB controller or are you just assigning the mouse using the web interface of unRAID itself. When you say the VM dropped off the network, do you mean the VM couldn't get out to the Internet or any local systems? Also, when was the last time you upgraded your VirtIO drivers inside the guest VM? Lastly, please try to collect diagnostics when these things happen and post them here. Without those diagnostics, we can't diagnose. Quote Link to comment
spencers Posted November 10, 2016 Share Posted November 10, 2016 After copying the VM disk from my array back to my SSD, the VM boots but is stuck at the windows 10 logo and spinning circle. I've turned off the GTX 1070 passthrough and just using VNC for simplicity's sake. I'm baffled as to what is wrong! The VM disk on the array boot just fine last night as a test. Why on earth won't it boot now after I copied it with midnight commander to my SSD? Make sure that you didn't miss any system or hidden files when you did the copy. (Some file managers don't automatically display/transfer those files.) I used midnight commander to copy the .qcow2 file from cache to array. Would that miss any files? Quote Link to comment
Frank1940 Posted November 10, 2016 Share Posted November 10, 2016 After copying the VM disk from my array back to my SSD, the VM boots but is stuck at the windows 10 logo and spinning circle. I've turned off the GTX 1070 passthrough and just using VNC for simplicity's sake. I'm baffled as to what is wrong! The VM disk on the array boot just fine last night as a test. Why on earth won't it boot now after I copied it with midnight commander to my SSD? Make sure that you didn't miss any system or hidden files when you did the copy. (Some file managers don't automatically display/transfer those files.) I used midnight commander to copy the .qcow2 file from cache to array. Would that miss any files? I am not sure. But if you use ls -al from the command line, it will show all of the files. Quote Link to comment
JorgeB Posted November 10, 2016 Share Posted November 10, 2016 I used midnight commander to copy the .qcow2 file from cache to array. Would that miss any files? I am not sure. But if you use ls -al from the command line, it will show all of the files. Vdisk is a single file, but I recommend using cp to keep the vdisk sparse, you'll lose that using MC. Quote Link to comment
spencers Posted November 11, 2016 Share Posted November 11, 2016 I used midnight commander to copy the .qcow2 file from cache to array. Would that miss any files? I am not sure. But if you use ls -al from the command line, it will show all of the files. Vdisk is a single file, but I recommend using cp to keep the vdisk sparse, you'll lose that using MC. Good to know. Thanks. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.